Galaxy Community Conference July 27, 2012 The National Center for Genome Analysis Support and Galaxy William K. Barnett, Ph.D. (Director) Richard LeDuc, Ph.D. (Manager) National Center for Genome Analysis Support

GCC. July 27, 2012National Center for Genome Analysis Support: Summary NCGAS and its mission NCGAS cyberinfrastructure The 100 Gigabit demonstration Scaling genomics analysis Trinity optimization

GCC. July 27, 2012National Center for Genome Analysis Support: Changing genomics analytical needs Next Gen sequencers are generating more data and getting cheaper Sequencing is:  Becoming commoditized at large centers and  Multiplying at individual labs Analytical capacity has not kept up  Bioinformatics support  Computational support (thousand points solution)  Storage support

GCC. July 27, 2012National Center for Genome Analysis Support: NCGAS widens the analytical bottleneck Funded by National Science Foundation Large memory clusters for assembly Bioinformatics consulting for biologists Optimized software for better efficiency Open for business at:

GCC. July 27, 2012National Center for Genome Analysis Support: Making it easier for Biologists Galaxy interface provides a “user friendly” window to NCGAS resources Supports many bioinformatics tools Available for both research and instruction. Common Rare Computational Skills LOW HIGH

Bio-IT World. April 25, 2012National Center for Genome Analysis Support: NCGAS Cyberinfrastructure at IU Mason large memory cluster (512 GB/node) Quarry cluster (16 GB/node) Data Capacitor (1 PB at 20 Gbps throughput) Research File System (RFS) for data storage Research Database Cluster for managing data sets. All interconnected with a high speed internal network (40 Gbps)

GCC. July 27, 2012National Center for Genome Analysis Support: GALAXY.IU.EDU Model Virtual box hosting Galaxy.IU.edu The host for each tool is configured to meet IU needs Quarry Mason Data Capacitor RFS UITS/NCGAS establishes tools, hardens them, and moves them into production. A custom Galaxy tool can be made to import data from the RFS to the DC. Individual labs can get duplicate boxes – provided they support it themselves. Policies on the DC guarantee that untouched data is removed with time.

GCC. July 27, 2012National Center for Genome Analysis Support: NCGAS Sandbox Demo at SC 11 STEP 1: data pre- processing, to evaluate and improve the quality of the input sequence STEP 2: sequence alignment to a known reference genome STEP 3: SNP detection to scan the alignment result for new polymorphisms

10 Gbps 100 Gbps NCGAS Mason (Free for NSF users) IU POD (12 cents per core hour) Amazon EC2 (20 cents per core hour) Data Capacitor NO data storage Charges Amazon Cloud Storage $80 – 120 per TB per month Lustre WAN File System Your Friendly Neighborhood Sequencing Center Your Friendly Neighborhood Sequencing Center Your Friendly Neighborhood Sequencing Center Two Options for Computation and Storage

GCC. July 27, 2012Customize footer: View menu/Header and Footer Commodity Internet (1Gbps but highly variable) Internet2 (100Gbps) Gbps NLR to Sequencing Centers (10Gbps/link) IU Data Capacitor (20 Gbps throughput) Ultra SCSI 160 Disk (1.2 Gbps, 160 MBps) DDR3 SDRAM (51.2 Gbps, 6.4GBps, ) This Architecture Scales!

GCC. July 27, 2012National Center for Genome Analysis Support: How would this work at scale? 1.Biologists use Galaxy to execute workflows 2.Sequence data mounted via Lustre WAN or automatically transferred using Internet2 3.Data Capacitor flows data into Mason or other computational clusters 4.Data Capacitor mounts or mirrors reference data from NCBI or other sources 5.Results delivered through web interfaces and to visualization or other science tools

Performance Improvements Richard LeDuc GCC /27/2012 Inchworm GraphFromFastA ReadsToTranscriptsQuantifyGraph Butterfly

Final Results Richard LeDuc GCC /27/2012

Trinity Results Significantly reduced runtime, while maintaining correctness of results Results are published Source code is commit to official SourceForge repository Continued support for HPC optimization for Trinity Brian Haas at Broad is developing Trinity workflows for Galaxy Richard LeDuc GCC /27/2012

Bio-IT World. April 25, 2012National Center for Genome Analysis Support: In Sum… NG Sequencing is creating a analytical problem that cannot be solved at sequencing centers NCGAS can provide a global scale infrastructure to better serve the needs of biologists who cannot become bioinformaticians to accomplish their research. Trinity is no longer a resource hog

GCC. July 27, 2012National Center for Genome Analysis Support: Thank You Questions? Bill Barnett Rich LeDuc